Author response: Sleep apnoea in patients undergoing colorectal cancer surgery: prospective cohort study

Correspondence to: Karl A. Franklin (email: karl.franklin@umu.se)

Department of Diagnostics and Intervention, Surgery

Umeå University

Umeå

Sweden

_____

BJS Open, https://doi.org/10.1093/bjsopen/zrag025, published 15 April 2026

_____

Thank you for giving us the opportunity to respond to the comments by Zhu and Zhuang. We value their arguments and we would like to clarify some points about screening and investigations of sleep apnoea before surgery.

We agree with doctors Zhu and Zhuang that our finding of moderate-to-severe sleep apnoea occurring in 48% of patients undergoing surgery for colorectal cancer is a wake-up call for perioperative physicians1.

We also agree that the low specificity of STOP-Bang > 3 of 42% is a significant barrier for its use in routine screening. Similar low specificities for STOP-Bang > 3 have also been reported in earlier studies2,3.

We, however, disagree with the suggested approach, using screening with STOP-Bang followed by home sleep apnoea testing for those scoring > 3. In our cohort, 139 of 198 (70%) of patients had a STOP-Bang score > 3, and the tool showed only moderate sensitivity (85%) and low specificity. Using STOP-Bang as a screening tool, as many as 70% of patients should then be suggested to undergo sleep apnoea investigations, still as many as 15% of patients with the disorder did not reach the limit for investigation.

STOP-Bang is a summary score of 8 questions symptoms and signs of sleep apnoea. However, most patients with sleep apnoea in the present study did not report snoring, witnessed apnoea or daytime sleepiness and few patients were younger than 50 years old. STOP-Bang also underestimate sleep apnoea in women since male sex includes 1 point, although sleep apnoea was as common in men as in women in the present cohort. Among the patients in this study, 85 % of all men scored > 3 on STOP-Bang, while only 43 % of women did, giving a sensitivity in women of 63 %. Although a STOP-Bang score with a threshold at 3 is suggested, the scores vary from 0-8 and the sensitivity is higher at low scores while the specificity is higher at high scores3. We therefore find screening with STOP-Bang problematic, and probably not necessary in patient undergoing colorectal cancer surgery since the prevalence of mild-to-severe sleep apnoea was as high as 79%. The question is rather to investigate the severity of sleep apnoea and home sleep apnoea testing would be the method of choice.

Given that sleep apnoea occurs in 79% of patients undergoing colorectal cancer surgery, and 48% have a moderatetosevere disease, we suggest simplified sleep apnoea recordings, i.e. level 3 sleep studies that monitor nasal airflow, pulse oximetry and respiratory effort, before surgery.